and ether. It melts at 224-5° C. to a reddish-brown oil. At a higher temperature it deflagrates, diffusing at the same time an aromatic odour. It is an almost perfectly indifferent substance. I could not even succeed in preparing a compound with dichloride of platinum; nitrate of silver, however, still gives, even in very dilute alcoholic solutions, a yellowish-green amorphous precipitate. When B-nitraniline is submitted to the action of nitrous acid, almost exactly the same phenomena are observed as those which occur in the preparation of the preceding compound. The crystals which have been deposited, but which in this case exhibit a perfectly distinct form, are easily purified in the same manner. The B-diazo-amidonitrobenzol differs only in a few points from the a compound. It is equally insoluble in water, and appears also to be as difficultly soluble in alcohol and ether. The two substances also seem to agree in their deportment with reagents, as far as could be decided by preliminary experiments. On the other hand, the melting point of B-diazo-amidonitrobenzol is 195.5° C., consequently 29° lower than that of a-diazo-amidonitrobenzol. The a compound crystallises as a rule in granular or moss-like shapes, whilst the B-diazo-amidonitrobenzol separates already during its preparation, in small though generally welldefined, ruby or reddish-yellow prisms, which by recrystallisation from alcohol or ether may be obtained of a considerable size. Diazo-amidodibromobenzol.-C12H,Br1N ̧ = CH2Br2N On passing a current of nitrous acid gas into a very dilute alcoholic solution of dibromaniline, the diazo-amidodibromobenzol is obtained as a bulky light-yellow precipitate. Repeated washing with alcohol renders it perfectly pure. Diazo-amidodibromobenzol has a great tendency to crystallise in different forms. From alcohol and ether, in which it is very difficultly soluble, even at the boiling temperature, it crystallises in fine golden-yellow interlaced needles, which melt at 167.5° C., but deflagrate at a higher temperature. On allowing an alcoholic VOL. XIX. F solution to evaporate spontaneously, yellowish-brown granules are frequently obtained which show a golden yellow fracture and a radiating crystalline structure. On several occasions, and under conditions not accurately determined, a-diazo-amidodibromobenzol, prepared from not absolutely pure dibromaniline crystallised from ether in extraordinarily beautiful yellow or ruby well-defined prisms. As this form, differing in such a characteristic manner from the yellow needles, appertained to the dibromaniline obtained from dibromisatin, I should have been inclined, in spite of what I had previously stated in regard to dibromaniline, to seek the cause of this difference in the dibromaniline prepared by the different methods. I soon, however, convinced myself, that the beautiful red crystals are converted, on further re-crystallisation from ether, into the same golden-yellow hair-like needles. is also prepared from dichloraniline and crystallises also in hairlike needles, which are, however, distinguished from the bromine compound by their light sulphur-yellow colour. It is insoluble in water, and very difficultly soluble in boiling alcohol and ether. It ́melts at 126·5° C. In their deportment with reagents, diazo-amidodibromobenzol and diazo-amidodichlorobenzol exhibit the greatest analogy to the nitrogen-substituted aniline-derivatives already described, whilst under the same conditions they yield corresponding products of decomposition. They also form precipitates with nitrate of silver. Their basic character, however, has completely disappeared; they no longer give platinum-salts; in fact, they possess more the character of an acid than that of a base, as they dissolve with ease in alcoholic potash, forming a reddish-brown solution, from which the original substance is precipitated in à perfectly unaltered state on the addition of an acid. It is, however, not possible to prepare salts of a definite composition. Aqueous potash has no action upon these bodies. From what has been stated it is obvious that the acidifying influence exerted by nitrogen, when taking the place of hydrogen in aniline, is repeated in an equal degree with the substitutionproducts of that body. If aniline and its bromine, hyponitric acid and other derivatives, be arranged in the order of their basic properties, there may be placed opposite to every individual of this series, a diazo double compound, the basic properties of which differ from the basic properties of the corresponding aniline body by the acidifying value of one atom of nitrogen. If the diazo-amidodibromobenzol corresponding to dibromaniline already exhibited distinct acid properties, it was to be expected that a body of a perfectly acid character would be obtained by the action of nitrous acid upon tribromaniline. I have instituted many experiments in order to obtain such a body, but without success. Nitrous acid has not, under any circumstances, the slightest action upon tribromaniline. Although it was to be expected that the researches conducted with the aniline group, described in the, preceding pages, would, when repeated with bases of analogous composition, lead to corresponding results, I have not omitted to prove this analogy by experiment. DIAZO-AMIDOTOLUOL.-C1H1N3 = (CH,NH,)} In order to prepare this body, toluidine (amidotoluol) is dissolved in a small quantity of strong alcohol and the solution is mixed with two or three times its volume of ether. After passing nitrous acid for a short time, the originally nearly colourless solution assumes a yellow colour. A drop of the solution is now taken out and evaporated on a watch-glass; if it leaves a residue of yellow needles, the operation is stopped, and the mixture of alcohol and ether is allowed to evaporate spontaneously. Yellow needle-shaped crystals are deposited, which must be washed with alcohol and then recrystallised from alcohol and ether. When dried over sulphuric acid and submitted to analysis, they gave numbers leading to the formula C14H15N3. Diazo-amidotoluol crystallises in yellow or reddish-yellow needles or prisms possessing a powerful lustre, and corresponding, in their solubility as well as in every other relation, with diazo-amidobenzol. Platinum-salt of Diazo-amidotoluol, C14H15N3.2HCl.2PtCl2 It is obtained in yellow glittering plates resembling iodide of lead, on mixing an alcoholic solution of diazo-amidotoluol with dichloride of platinum. It deflagrates at a high temperature. This body falls in small yellow crystals on passing nitrous acid into a dilute alcoholic solution of nitranisidine* [amidonitranisol = C,H6(NO2) (NH2)O]. It forms microscopic yellow needles, insoluble n water, difficultly soluble in hot alcohol and ether. When dry they are rendered highly electric by friction. They melt, on the application of heat, to a reddish-brown oil, and deflagrate on raising the temperature. Diazo-amidonitranisol, under the influence of reagents, appears to undergo changes similar to those of its representatives in the aniline group. On referring once more to the most characteristic points of the nitrogen substitution-products of the hydrocarbons of the aromatic series, it will be observed that, not only in their formation, but also in their properties and decompositions, they correspond in every relation with the peculiarities of that series of acids of which the best studied individual is diazo-amidobenzoic acid. This is but natural, as all the transformations which benzoic acid and its homologues undergo, under the influence of substitution processes, are also effected in the hydrocarbons which in combination with CO2 form the corresponding aromatic acids. In the same measure as benzoic acid on distillation with baryta yields carbonic acid and benzol, or as amidobenzoic acid splits up into carbonic acid and amidobenzol, so, doubtless, would diazoamidobenzol be obtained from diazo-amidobenzoic acid under similar conditions, if the instability of this compound did not preclude every possibility of its formation under such circumstances. * The nitranisidine was obtained from dinitranisol, which was prepared according to Cahour's method from nitranisic acid. The following compounds stand in the same relation :— In a preliminary note on these bodies I referred them to the type Ho H2 N2. This was done at a time when I knew but little of their H2 products of decomposition. If, after the observations communicated in this paper, diazo-amidobenzol, for instance, were still to be considered as a diamine, having the formula C6H5 CH N" H endeavouring to explain the phenomena of its decompositions, the same inconsistencies would become apparent, as when supposing diazo-amidobenzoic acid to be constituted according to the rational formula [(C,H ̧O)′′HN′′]ˇN1⁄2 02 H2 IX.-Action of Heat on Ferric Hydrate in presence of Water. By EDWARD DAVIES, F.C.S., Liverpool. HAVING noticed the occurrence of shining particles, apparently crystalline, in precipitated ferric oxide which had been boiled for a considerable time, I made the following experiments with a view to ascertain if the oxide could be obtained in the anhydrous state by boiling with water. Being aware that M. Péan de St. Gilles has investigated the properties of the oxide thus altered, I confined myself to the determination of the amount of water. In the memoir above cited the oxide is spoken of as being a hydrate; it appears, however, that it is for the most part anhydrous. In the following experiments the ferric hydrate was prepared from ferric VOL. XIX. G |